summaryrefslogtreecommitdiffstats
path: root/doctor_tests/scenario/maintenance.py
blob: 2e40529f23cd47821bd4dbd96b2c894551f1cd8d (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
##############################################################################
# Copyright (c) 2019 Nokia Corporation and others.
#
# All rights reserved. This program and the accompanying materials
# are made available under the terms of the Apache License, Version 2.0
# which accompanies this distribution, and is available at
# http://www.apache.org/licenses/LICENSE-2.0
##############################################################################
import datetime
import json
import requests
import time

from doctor_tests.admin_tool import get_admin_tool
from doctor_tests.app_manager import get_app_manager
from doctor_tests.common.utils import get_doctor_test_root_dir
from doctor_tests.identity_auth import get_identity_auth
from doctor_tests.identity_auth import get_session
from doctor_tests.inspector import get_inspector
from doctor_tests.os_clients import keystone_client
from doctor_tests.os_clients import neutron_client
from doctor_tests.os_clients import nova_client
from doctor_tests.stack import Stack


class Maintenance(object):

    def __init__(self, trasport_url, conf, log):
        self.conf = conf
        self.log = log
        self.admin_session = get_session()
        self.keystone = keystone_client(
            self.conf.keystone_version, get_session())
        self.nova = nova_client(conf.nova_version, get_session())
        auth = get_identity_auth(project=self.conf.doctor_project)
        self.neutron = neutron_client(get_session(auth=auth))
        self.stack = Stack(self.conf, self.log)
        if self.conf.admin_tool.type == 'sample':
            self.admin_tool = get_admin_tool(trasport_url, self.conf, self.log)
            self.endpoint = 'maintenance'
        else:
            self.endpoint = 'v1/maintenance'
        self.app_manager = get_app_manager(self.stack, self.conf, self.log)
        self.inspector = get_inspector(self.conf, self.log, trasport_url)

    def get_external_network(self):
        ext_net = None
        networks = self.neutron.list_networks()['networks']
        for network in networks:
            if network['router:external']:
                ext_net = network['name']
                break
        if ext_net is None:
            raise Exception("external network not defined")
        return ext_net

    def setup_maintenance(self, user):
        # each hypervisor needs to have same amount of vcpus and they
        # need to be free before test
        hvisors = self.nova.hypervisors.list(detailed=True)
        prev_vcpus = 0
        prev_hostname = ''
        self.log.info('checking hypervisors.......')
        for hvisor in hvisors:
            vcpus = hvisor.__getattr__('vcpus')
            vcpus_used = hvisor.__getattr__('vcpus_used')
            hostname = hvisor.__getattr__('hypervisor_hostname')
            if vcpus < 2:
                raise Exception('not enough vcpus (%d) on %s' %
                                (vcpus, hostname))
            if vcpus_used > 0:
                if self.conf.test_case == 'all':
                    # VCPU might not yet be free after fault_management test
                    self.log.info('%d vcpus used on %s, retry...'
                                  % (vcpus_used, hostname))
                    time.sleep(15)
                    hvisor = self.nova.hypervisors.get(hvisor.id)
                    vcpus_used = hvisor.__getattr__('vcpus_used')
                if vcpus_used > 0:
                    raise Exception('%d vcpus used on %s'
                                    % (vcpus_used, hostname))
            if prev_vcpus != 0 and prev_vcpus != vcpus:
                raise Exception('%d vcpus on %s does not match to'
                                '%d on %s'
                                % (vcpus, hostname,
                                   prev_vcpus, prev_hostname))
            prev_vcpus = vcpus
            prev_hostname = hostname

        # maintenance flavor made so that 2 instances take whole node
        flavor_vcpus = int(vcpus / 2)
        compute_nodes = len(hvisors)
        amount_actstdby_instances = 2
        amount_noredundancy_instances = 2 * compute_nodes - 2
        self.log.info('testing %d computes with %d vcpus each'
                      % (compute_nodes, vcpus))
        self.log.info('testing %d actstdby and %d noredundancy instances'
                      % (amount_actstdby_instances,
                         amount_noredundancy_instances))
        max_instances = (amount_actstdby_instances +
                         amount_noredundancy_instances)
        max_cores = compute_nodes * vcpus

        user.update_quota(max_instances, max_cores)

        test_dir = get_doctor_test_root_dir()
        template_file = '{0}/{1}'.format(test_dir, 'maintenance_hot_tpl.yaml')
        files, template = self.stack.get_hot_tpl(template_file)

        ext_net = self.get_external_network()

        parameters = {'ext_net': ext_net,
                      'flavor_vcpus': flavor_vcpus,
                      'maint_image': self.conf.image_name,
                      'nonha_intances': amount_noredundancy_instances,
                      'ha_intances': amount_actstdby_instances}

        self.log.info('creating maintenance stack.......')
        self.log.info('parameters: %s' % parameters)

        self.stack.create('doctor_test_maintenance',
                          template,
                          parameters=parameters,
                          files=files)

        if self.conf.admin_tool.type == 'sample':
            self.admin_tool.start()
        else:
            # TBD Now we expect Fenix is running in self.conf.admin_tool.port
            pass
        self.app_manager.start()
        self.inspector.start()

    def start_maintenance(self):
        self.log.info('start maintenance.......')
        hvisors = self.nova.hypervisors.list(detailed=True)
        maintenance_hosts = list()
        for hvisor in hvisors:
            hostname = hvisor.__getattr__('hypervisor_hostname')
            maintenance_hosts.append(hostname)

        url = ('http://%s:%s/%s' %
               (self.conf.admin_tool.ip,
                self.conf.admin_tool.port,
                self.endpoint))
        headers = {
            'Content-Type': 'application/json',
            'Accept': 'application/json'}
        if self.conf.admin_tool.type == 'fenix':
            headers['X-Auth-Token'] = self.admin_session.get_token()
        self.log.info('headers %s' % headers)
        retries = 12
        ret = None
        while retries > 0:
            # let's start maintenance 20sec from now, so projects will have
            # time to ACK to it before that
            maintenance_at = (datetime.datetime.utcnow() +
                              datetime.timedelta(seconds=30)
                              ).strftime('%Y-%m-%d %H:%M:%S')

            data = {'state': 'MAINTENANCE',
                    'maintenance_at': maintenance_at,
                    'metadata': {'openstack_version': 'Rocky'},
                    'workflow': 'default'}

            if self.conf.admin_tool.type == 'sample':
                data['hosts'] = maintenance_hosts
            else:
                data['hosts'] = []
            try:
                ret = requests.post(url, data=json.dumps(data),
                                    headers=headers)
            except:
                if retries == 0:
                    raise Exception('admin tool did not respond in 120s')
                else:
                    self.log.info('admin tool not ready, retry in 10s')
                retries = retries - 1
                time.sleep(10)
                continue
            break
        if not ret:
            raise Exception("admin tool did not respond")
        if ret.status_code != 200:
            raise Exception(ret.text)
        return ret.json()['session_id']

    def remove_maintenance_session(self, session_id):
        self.log.info('remove maintenance session %s.......' % session_id)
        url = ('http://%s:%s/%s/%s' %
               (self.conf.admin_tool.ip,
                self.conf.admin_tool.port,
                self.endpoint,
                session_id))

        headers = {
            'Content-Type': 'application/json',
            'Accept': 'application/json'}

        if self.conf.admin_tool.type == 'fenix':
            headers['X-Auth-Token'] = self.admin_session.get_token()

        ret = requests.delete(url, data=None, headers=headers)
        if ret.status_code != 200:
            raise Exception(ret.text)

    def get_maintenance_state(self, session_id):
        url = ('http://%s:%s/%s/%s' %
               (self.conf.admin_tool.ip,
                self.conf.admin_tool.port,
                self.endpoint,
                session_id))

        headers = {
            'Content-Type': 'application/json',
            'Accept': 'application/json'}

        if self.conf.admin_tool.type == 'fenix':
            headers['X-Auth-Token'] = self.admin_session.get_token()

        ret = requests.get(url, data=None, headers=headers)
        if ret.status_code != 200:
            raise Exception(ret.text)
        return ret.json()['state']

    def wait_maintenance_complete(self, session_id):
        retries = 90
        state = None
        time.sleep(300)
        while (state not in ['MAINTENANCE_DONE', 'MAINTENANCE_FAILED'] and
               retries > 0):
            time.sleep(10)
            state = self.get_maintenance_state(session_id)
            retries = retries - 1
        self.remove_maintenance_session(session_id)
        self.log.info('maintenance %s ended with state %s' %
                      (session_id, state))
        if state == 'MAINTENANCE_FAILED':
            raise Exception('maintenance %s failed' % session_id)
        elif retries == 0:
            raise Exception('maintenance %s not completed within 20min' %
                            session_id)

    def cleanup_maintenance(self):
        if self.conf.admin_tool.type == 'sample':
            self.admin_tool.stop()
        self.app_manager.stop()
        self.inspector.stop()
        self.log.info('stack delete start.......')
        self.stack.delete()